[go: up one dir, main page]

CN109495778A - film editing method, device and system - Google Patents

film editing method, device and system Download PDF

Info

Publication number
CN109495778A
CN109495778A CN201710874533.6A CN201710874533A CN109495778A CN 109495778 A CN109495778 A CN 109495778A CN 201710874533 A CN201710874533 A CN 201710874533A CN 109495778 A CN109495778 A CN 109495778A
Authority
CN
China
Prior art keywords
picture frame
film
new
frame
picture
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201710874533.6A
Other languages
Chinese (zh)
Other versions
CN109495778B (en
Inventor
谢佳达
卓卿安
刘政德
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wistron Corp
Original Assignee
Wistron Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Wistron Corp filed Critical Wistron Corp
Publication of CN109495778A publication Critical patent/CN109495778A/en
Application granted granted Critical
Publication of CN109495778B publication Critical patent/CN109495778B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/431Generation of visual interfaces for content selection or interaction; Content or additional data rendering
    • H04N21/4318Generation of visual interfaces for content selection or interaction; Content or additional data rendering by altering the content in the rendering process, e.g. blanking, blurring or masking an image region
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B27/00Editing; Indexing; Addressing; Timing or synchronising; Monitoring; Measuring tape travel
    • G11B27/02Editing, e.g. varying the order of information signals recorded on, or reproduced from, record carriers
    • G11B27/031Electronic editing of digitised analogue information signals, e.g. audio or video signals
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/85Assembly of content; Generation of multimedia applications
    • H04N21/854Content authoring
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/23Processing of content or additional data; Elementary server operations; Server middleware
    • H04N21/234Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs
    • H04N21/23418Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/23Processing of content or additional data; Elementary server operations; Server middleware
    • H04N21/234Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs
    • H04N21/2343Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving reformatting operations of video signals for distribution or compliance with end-user requests or end-user device requirements
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/431Generation of visual interfaces for content selection or interaction; Content or additional data rendering
    • H04N21/4312Generation of visual interfaces for content selection or interaction; Content or additional data rendering involving specific graphical features, e.g. screen layout, special fonts or colors, blinking icons, highlights or animations
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/432Content retrieval operation from a local storage medium, e.g. hard-disk
    • H04N21/4325Content retrieval operation from a local storage medium, e.g. hard-disk by playing back content from the storage medium

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Databases & Information Systems (AREA)
  • Television Signal Processing For Recording (AREA)
  • Studio Devices (AREA)

Abstract

A method, a device and a system for editing a film are provided. The target is obtained from the original movie. Based on a plurality of image feature points of each picture frame, obtaining the relative position relation between each picture frame and two picture frames adjacent to the picture frame in front and at the back. And respectively executing an adjusting program on the picture frames. The adjustment procedure comprises the following steps: taking out a picture frame N; identifying a target in the frame N; adjusting the frame N to obtain a new frame so that the target is located at the center of the new frame; and modifying the new frame based on the relative position relationship to obtain a post-frame. Outputting the processed movie based on the plurality of post-drawing frames.

Description

Film edit methods, apparatus and system
Technical field
The present invention relates to a kind of film processing methods, apparatus and system, and in particular to a kind of film edit methods, device And system.
Background technique
The function of recorded film has been indispensable function on mobile phone and camera, but the big portion of quality of film quality Divide the technology depending on user.Such as the mobile too fast or too slow, person of being taken of mobile phone or camera is always in film center etc., It is all result caused by user's technology is bad.Especially photographer and the person of being taken all move simultaneously with clap in the case of, more The film that difficulty has been taken.And the current rear software made usually only simply captures or the functions such as trimming, can not modify film Content.Therefore the bad film of a quality, also can only obtain film after the bad trimming of quality, does not do after pruned Method changes film quality.
Summary of the invention
The present invention provides a kind of film edit methods, apparatus and system, and the target position in film can be allowed to exist after processing Picture center, promotes film quality.
Film edit methods of the invention, include the following steps.Original film is received, wherein original film includes multiple figures Frame.Target is obtained from original film.Multiple image feature points based on each picture frame, obtain each picture frame respectively with its front and back phase Relative positional relationship between two adjacent picture frames.Adjustment programme is executed to the picture frame respectively.Adjustment programme includes following steps It is rapid: picture frame N is taken out from the picture frame;Recognize the target in picture frame N;Picture frame N is adjusted to obtain new picture frame, so that target is located at The center of new picture frame;And new picture frame is modified based on relative positional relationship and then the frame that charts after obtaining.Based on multiple rear drawing Frame come export processing after film.
In one embodiment of this invention, picture frame N is adjusted to obtain new picture frame, so that target is located at the center of new picture frame Step includes: that the size of foundation picture frame N establishes blank picture frame;And by being the center of blank picture frame by goal-setting, and incite somebody to action Picture frame N coincides to blank picture frame and obtains new picture frame.
In one embodiment of this invention, charted frame after being modified new picture frame based on relative positional relationship and then being obtained Before step, further includes: judge whether new picture frame complete, in the incomplete situation of new picture frame, based on relative positional relationship come Modify new picture frame and then the frame that charts after obtaining;And in the complete situation of new picture frame, new picture frame is not modified and stores new picture frame For rear drawing frame.
In one embodiment of this invention, in the incomplete situation of new picture frame, white space institute in new picture frame is detected Position;One close to the position is taken out from the corresponding picture frame N-1 of picture frame N and picture frame N+1 based on relative positional relationship It is a to be used as compensation picture frame;And based on new picture frame and compensation picture frame, image after being adjusted.
In one embodiment of this invention, after the position where in detection white space in new picture frame, further includes: base Position where white space is in new picture frame sets reference side;Judge whether picture frame N-1 and picture frame N+1 are located at picture frame N's Reference side;If determining that picture frame N-1 or picture frame N+1 is located at the reference side of picture frame N, based on relative positional relationship from picture frame N-1 and One that close reference side is taken out in picture frame N+1 is used as compensation picture frame;And if determining that picture frame N-1 and picture frame N+1 neither exist The reference side of picture frame N then abandons adjustment picture frame N.
In one embodiment of this invention, based on new picture frame and compensation picture frame, after being adjusted the step of image include: According to the size and location of image feature point, white space, block corresponding with white space is cut from compensation picture frame, it will be described Block conforms to the white space of new picture frame.
In one embodiment of this invention, based on new picture frame and compensation picture frame, after being adjusted the step of image include: Based on image feature point, new picture frame is coincided on compensation picture frame and obtains the image that coincides;And it is cut centered on the target Under coincide block identical with the size of new picture frame in image.
In one embodiment of this invention, it is also wrapped after being adjusted after image based on new picture frame and compensation picture frame Include: whether image is complete after judging adjustment;If image is complete after adjustment, then stores after image after adjustment is and chart frame;And If image is imperfect after adjustment, then another picture frame different from picture frame N adjacent with compensation picture frame is taken out, and be based on another figure Frame continues to modify.
In one embodiment of this invention, the image feature point based on each picture frame, obtains each picture frame respectively and before it The step of relative positional relationship between two adjacent picture frames includes: based on the image feature point and picture frame N+1 in picture frame N afterwards In image feature point, judge first relative position of the picture frame N+1 relative to picture frame N;Based in picture frame N image feature point with Image feature point in picture frame N-1, judges second relative position of the picture frame N-1 relative to picture frame N;And the opposite position of storage first Set the relative positional relationship for being picture frame N with the second relative position.
Film editing device of the invention, comprising: storage device and processor.Storage device stores multiple modules, Including Object selection module, film pre-processing module, film modification module and film output module.Processor is coupled to storage Device executes the module to edit to original film.Original film includes multiple picture frames.Wherein, Object selection module Receiving includes original film, and obtains target from original film.Multiple image features of the film pre-processing module based on each picture frame Point obtains relative positional relationship of each picture frame respectively between its adjacent two picture frame in front and back.It is right respectively that film modifies module The picture frame executes adjustment programme.Adjustment programme includes: the taking-up picture frame N from the picture frame;Recognize the target in picture frame N;It adjusts Whole picture frame N is to obtain new picture frame, so that target is located at the center of new picture frame;And new picture frame is modified based on relative positional relationship And then the frame that charts after obtaining.Film output module exports film after processing based on multiple rear drawing frames.
Film editing system of the invention, comprising: calculation server and electronic device.Electronic device includes first network Chip, the first storage device and first processor.First storage device stores Object selection module.First processor is coupled to First storage device, performance objective choose module, receive the original film including multiple picture frames by Object selection module, and Target is obtained from original film, later, original film and target are transmitted to calculation server by first network chip, so that Calculation server carries out film editor to original film.Calculation server include the second network chip, the second storage device and Second processor.Second network chip receives original film and target from electronic device by network.The storage of second storage device There are multiple modules, including film pre-processing module, film modification module and film output module.Second processor is coupled to Storage device drives the module to carry out film editor.Multiple images of the film pre-processing module based on each picture frame are special Point is levied, relative positional relationship of each picture frame respectively between its adjacent two picture frame in front and back is obtained.Film modifies module difference Adjustment programme is executed to the picture frame.Adjustment programme includes: the taking-up picture frame N from the picture frame;Recognize the target in picture frame N; Picture frame N is adjusted to obtain new picture frame, so that target is located at the center of new picture frame;And new figure is modified based on relative positional relationship Chart frame after frame and then acquisition.Film output module exports film after processing based on multiple rear drawing frames.
Based on above-mentioned, in selection film target, and target is placed in the center of picture frame and is carried out using adjacent frames Compensation, therefore, the problem of target offset center is caused when can correct film shooting person's moving lens.
To make the foregoing features and advantages of the present invention clearer and more comprehensible, special embodiment below, and it is detailed to cooperate attached drawing to make Carefully it is described as follows.
Detailed description of the invention
Fig. 1 is the block diagram according to the film editing device of one embodiment of the invention.
Fig. 2A and Fig. 2 B is the flow chart according to the film edit methods of one embodiment of the invention.
Fig. 3 is the schematic diagram according to the judgement relative positional relationship of one embodiment of the invention.
Fig. 4 is the schematic diagram for establishing new picture frame according to one embodiment of the invention.
Fig. 5 is the method flow diagram according to the modification picture frame of one embodiment of the invention.
Fig. 6 A~Fig. 6 D is the schematic diagram according to the modification picture frame of one embodiment of the invention.
Fig. 7 is the block diagram according to the film editing system of one embodiment of the invention.
Description of symbols:
100: film editing device
110,712,722: processor
120,713,723: storage device
121,714: Object selection module
123,724: film pre-processing module
125,725: film modifies module
127,726: film output module
601, N_blank: white space
603: region
610,620, N, N+1: picture frame
630, N_new: new picture frame
640: image after adjustment
700: film editing system
710: electronic device
720: calculation server
711,721: network chip
A, A ': image feature point
T: target
S205~S225, S220_1~S220_7: each step of film edit methods
S510~S560: each step of method of picture frame is modified
Specific embodiment
Fig. 1 is the block diagram according to the film editing device of one embodiment of the invention.Please refer to Fig. 1, film editing device 100 include processor 110 and storage device 120.And storage device 120 includes Object selection module 121, film pre-treatment mould Block 123, film modification module 125 and film output module 127.
Film editing device 100 is, for example, smart phone, desktop computer, laptop, tablet computer, server, intelligence Intelligent TV etc. has the electronic device of operational capability.Processor 110 is, for example, central processing unit (Central Processing Unit, CPU), image processing unit (Graphic Processing Unit, GPU), physical processing unit (Physics Processing Unit, PPU), programmable microprocessor (Microprocessor), embedded control chip, digital signal Processor (Digital Signal Processor, DSP), special application integrated circuit (Application Specific Integrated Circuits, ASIC) or other similar device.Storage device 120 is, for example, the fixed of any pattern or can Mobile random access memory (Random Access Memory, RAM), read-only memory (Read-Only Memory, ROM), flash memory (Flash memory), safe digital card (Secure Digital Memory Card, SD), hard disk Or the combination of other similar device or these devices.
Processor 110 is coupled to storage device 120, and drives Object selection module 121, film pre-processing module 123, shadow Piece modification module 125 and film output module 127 edit original film to execute.
Fig. 2A and Fig. 2 B is the flow chart according to the film edit methods of one embodiment of the invention.Referring to Fig.1, Fig. 2A and figure 2B receives original film by Object selection module 121 in step S205.Original film includes multiple picture frames (frame).Also, Object selection module 121 can obtain the size of original film after receiving original film.For example, shadow Piece editing device 100 by universal serial bus (Universal Serial Bus, USB), cable network, wireless network (with Network (Ethernet) or original film WiFi) etc. is received from photographic device very much, so, do not limit coming for original film herein Source.
Then, in step S210, Object selection module 121 obtains target from original film.Here, Object selection module 121 offer frames select function, and selecting function by frame, a picture frame center selects target wherein.For example, film editing device 100 include display and input unit.Original film is shown in display to play out by Object selection module 121, is made User suspends broadcasting by input unit (for example, slide-mouse, Touch Screen), is played work as in pause using input unit later Carry out frame in lower shown picture frame and selects target.
After obtaining target, Object selection module 121 will analyze this target to obtain its corresponding target signature. Such as using open source computer image vision function library (Open Source Computer Vision Library, OpenCV histogram (histogram) algorithm in) or the algorithm of profile (contour) is found to obtain texture, face The features such as color, profile.After obtaining target signature, Object selection module 121 can store target signature, to make to subsequent With.
Later, multiple image features in step S215, by film pre-processing module 123 based on each picture frame Point obtains relative positional relationship of each picture frame respectively between its adjacent two picture frame in front and back.
Specifically, film pre-processing module 123 can convert (Scale-invariant using scale invariant feature Feature transform, SIFT) algorithm or accelerate robust feature (Speeded Up Robust Features, SURF) acquisition of the characteristic points such as algorithm and feature point description method, to obtain multiple image feature points in each picture frame, into And store the coordinate and the characteristic direction of each image feature point, characteristic strength, dimension difference of each image feature point. Later, relative position of each picture frame respectively between its two neighboring picture frame in front and back is obtained using these image feature points Relationship, and store image feature point and relative positional relationship.That is, each picture frame can be recorded for each picture frame Relative positional relationship between previous picture frame and next picture frame adjacent thereto respectively.
Here, only judging that its relative position with picture frame N-1 and its between picture frame N+1 is closed when handling picture frame N System.That is, film pre-processing module 123 based in picture frame N image feature point and picture frame N+1 in image feature point, Judge first relative position of the picture frame N+1 relative to picture frame N.Also, film pre-processing module 123 is based on the image in picture frame N Image feature point in characteristic point and picture frame N-1, judges second relative position of the picture frame N-1 relative to picture frame N.Later, it stores The relative positional relationship that first relative position and the second relative position are picture frame N.
For convenience of explanation, beneath that relative positional relationship is judged with left and right displacement.So, in other embodiments, also may be used The relative positional relationship that upper and lower displacement is judged with similar gimmick, is not limited thereto.
Firstly, all image feature points for taking out the one of image feature point A and picture frame N+1 of picture frame N are compared, To find out the point to match in picture frame N+1 with image feature point A.In the institute of all image feature points and picture frame N+1 to picture frame N There is image feature point to carry out judging picture frame N+1 one by one with the coordinates of the two image feature points to match after matching terminates Positioned at the left or right side of picture frame N.Finally, the relative position of each group image characteristic point after statistical match, judge whereby Picture frame N+1 is located at the quantity of the left or right side of picture frame N, represents relative position with side more than quantity.
Fig. 3 is the schematic diagram according to the judgement relative positional relationship of one embodiment of the invention.Referring to figure 3., in time sequence On column, the picture frame of the previous acquisition of picture frame N+1 is picture frame N, the image feature point of the image feature point A and picture frame N+1 of picture frame N A ' matches.It in the present embodiment, is to get off to be illustrated the case where moving left and right and (move horizontally) with camera lens.Picture frame N+1's The coordinate of image feature point A ' is located at the right side of the coordinate of the image feature point A of picture frame N, that is, the horizontal of image feature point A ' sits The horizontal coordinate of image feature point A of the mark greater than picture frame N.Accordingly, indicate that camera lens is turned left follow shot when shooting original film, It therefore include the content that left side is more leaned on than content shown by picture frame N in picture frame N+1.Therefore for determining with image feature point A, Picture frame N+1 is located at the left side of picture frame N, is stored as N+1 { point A;L=1;R=0 }, wherein L represents left side, and R represents right side. And so on, the relative positional relationship of each group image feature point to match between picture frame N and picture frame N+1 can be found out, also, The relative positional relationship of each group image feature point to match between picture frame N and picture frame N-1 can be found out.
For example, the image feature point A " (not being painted) to match with the image feature of picture frame N point A is found out in picture frame N-1, Judge that the coordinate of image feature point A is located at the right side (such as camera lens is turned right and shot) of image feature point A ", is then stored as N-1 {point A;L=0;R=1 }.That is, before storage can be corresponded at each image feature point of each picture frame with it The relative position of two neighboring picture frame afterwards.
In addition, in another embodiment, being compared in the case where picture frame N+1 and picture frame N-1 are located at the same side of picture frame N One (for example, picture frame N+1) positioned at the leftmost side is found out, and its corresponding position relationship is stored as N+1 { point A;L=1; R=0 }, N-1 { point A;L=0;R=0 }.
In addition, in other embodiments, in the case where camera lens is to move up and down and (vertically move), image feature point A's Relative positional relationship recording mode is as follows: if picture frame N+1 is located at the top of picture frame N, being stored as N+1 { point A;U=1;D= 0};If picture frame N+1 is located at the lower section of picture frame N, it is stored as N+1 { point A;U=0;D=1 }, wherein U represents top, and D is represented Lower section.
After having compared the image feature point between picture frame N and picture frame N+1 and picture frame N-1, find out three picture frames all The image feature point occurred is counted, and finds out the left or right side that picture frame N+1 or picture frame N-1 is located at picture frame N.For example, The image feature for appearing in picture frame N, picture frame N+1 and picture frame N-1 simultaneously is A, B, C, if the parameter corresponding to picture frame N-1 is {point A;L=1;R=0 }, { point B;L=0;R=0 }, { point C;L=1;R=0 }, indicate the shadow for being greater than half As characteristic point (A, C) all ballots think picture frame N-1 in the left side of picture frame N, that is, after counting each common image feature point, Judge that picture frame N-1 is more in the image feature point in the left side of picture frame N.Accordingly, just determine that picture frame N-1 is located at the left side of picture frame N.With This analogizes, and picture frame N+1 judges in the same manner.
Fig. 2A is returned to, in step S220, film modifies module 125 and executes adjustment programme to the picture frame respectively.Here, Adjustment programme is as shown in Figure 2 B, including step S220_1~S220_7.
In step S220_1, picture frame N is taken out.Then, in step S220_3, the target in picture frame N is recognized.For example, A picture frame N is sequentially taken out since Fig. 1 frame, wherein N is greater than or equal to 1.Film modification module 125 can utilize mean vector It deviates (mean shift) algorithm or supports vector machine (Support Vector Machine, SVM) and be based on target signature To find out target.Object selection module 121 analyze from original film obtain target and after obtaining its corresponding target signature, Target signature can be stored, so that film modification module 125 uses.
In step S220_5, picture frame N is adjusted to obtain new picture frame, so that target is located at the center of new picture frame.Citing comes It says, film modification module 125 establishes blank picture frame according to the size of picture frame N;And by being blank picture frame by goal-setting Center, picture frame N is coincided to blank picture frame and obtains new picture frame.Or it first coincides with blank picture frame, then move during target is The heart, herein not restriction sequence.
Fig. 4 is the schematic diagram for establishing new picture frame according to one embodiment of the invention.Referring to figure 4., picture frame N is adjusted, it will New image N_new is obtained centered on target T setting.Here, being drawn with a dashed line to indicate new image N_new.And target T not Position can have white space N_blank in the new image N_new of acquisition in the case where the center of picture frame N.In the present embodiment In, for target T at the side to the right of picture frame N, then the new image N_new obtained can have white space N_blank on right side.And In other embodiments, if the new image N_new that target T at the side to the left of picture frame N, is then obtained can have blank area in left side Domain.
Fig. 2 B is returned, in step S220_7, new picture frame is modified based on relative positional relationship and then the frame that charts after obtaining. It is to select the figure adjacent with picture frame N in which position of new picture frame that film, which modifies module 125 according to the white space in new picture frame, Frame N+1 or picture frame N-1 modifies new picture frame, and the frame that charts after obtaining.For example, white space is located at left side, then in picture frame N+1 and It is taken out in picture frame N-1 and is located at picture frame N left side person to modify new picture frame;White space is located at right side, then in picture frame N+1 and picture frame N- It is taken out in 1 and is located at picture frame N right side person to modify new picture frame.In another embodiment, rightmost side picture frame or leftmost diagram are selected Frame is modified.For example, being coincided downwards using image feature point, or change after can also scaling or rotate according to image feature point It closes.In addition, database after can also further storing the rear frame that charts obtained to a modification.
Finally, returning to Fig. 2A, in step S225, film output module 127 charts frame to export shadow after processing after being based on Piece.For example, film output module 127 can read all rear drawing frames in database from after modifying, in conjunction with all rear frames that chart And film after being handled.
The process of beneath for example bright modification picture frame.Fig. 5 is the method according to the modification picture frame of one embodiment of the invention Flow chart.Referring to figure 5., after step S220_5, film modifies module 125 and executes step S510, whether judges new picture frame Completely.For example, judging that new picture frame whether there is white space, or judge whether new picture frame has defect area.In new picture frame In there are when white space or defect area, determine that new picture frame is imperfect.And in the case where determining the new incomplete situation of picture frame, it holds Row step S220_7.Step S220_7 further comprises step S520~S555.
On the other hand, in the complete situation of new picture frame, such as there is no white space or defect areas, such as step Shown in S515, film modification module 125 does not modify new picture frame and directly stores the frame that charts after new picture frame is.That is, film modifies mould Block 125 (center that expression target script is just located at picture frame N) in the case where determining that new picture frame is complete situation, directly can newly scheme in storage Database is without modifying new picture frame after frame to modification.
In the new picture frame of judgement there are in the case where white space, position of the white space where in new picture frame is further detected It sets, one for taking out the close position from picture frame N-1 and picture frame N+1 based on relative positional relationship is used as compensation picture frame.
Specifically, film modifies module 125 and reads the picture frame N-1 and picture frame N+ for corresponding to picture frame N in step S520 1 respective relative positional relationship.Also, film modify module 125 detection white space in new picture frame where position it Afterwards, the position where based on white space in new picture frame sets reference side.For example, in white space in the left side of new picture frame When, reference side is set as left side;In white space when the right side of new picture frame, reference side is set as right side.
Then, in step S525, judge whether picture frame N+1 or picture frame N-1 are located at the reference side of picture frame N.If determining Picture frame N-1 and picture frame N+1 are neither in the reference side of picture frame N, as shown in step S530, abandon adjustment picture frame N.
And it is located at the reference side of picture frame N in judgement picture frame N+1 or picture frame N-1, in step S535, it is based on step S520 institute The relative positional relationship of acquisition takes out the picture frame close to reference side from picture frame N-1 and picture frame N+1 as compensation picture frame.For example, Assuming that reference side is left side, and it includes N+1 { point A that the relative position taken out, which is closed,;L=1;R=0 } and N-1 { point A;L =0;R=1 }, since picture frame N+1 is close to left side, using picture frame N+1 as compensation picture frame.In another example, it is assumed that reference side is Top, and it includes N+1 { point A that the relative position taken out, which is closed,;U=1;D=0 } and N-1 { point A;U=0;D=1 }, by In picture frame N+1 be by close over, therefore using picture frame N+1 as compensate picture frame.
Then, in step S540, image after being adjusted based on compensation picture frame and new picture frame.Here, can be used " Clip and paste " mode, that is, according to the size and location of white space, be based on image feature point, it is corresponding from compensation picture frame Block identical with the size of white space is cut at position, and the block is conformed into the white space of new picture frame to become and adjust Image after whole.Alternatively, it is also possible to use the mode of " piecing together ", that is, be based on image feature point, new picture frame is coincided in compensation picture frame Upper and acquisition coincides in image, then the image that coincides certainly, is cut in a manner of keeping the target of new picture frame centrally located and new picture frame The identical block of size, and image after being adjusted.Namely first new picture frame is covered on compensation picture frame, then cuts phase again With image after the adjustment of size.The mode of " piecing together " needs to position the image boundary of image after the adjustment to be cut.Position boundary Method can recorde the boundary of new picture frame and redefine side as the boundary cut, or centered on the target of new picture frame Boundary.
Coincide new picture frame and compensation picture frame.The method to coincide can find identical according to SIFT algorithm or SURF algorithm Image feature point coincides according to identical image feature point.Here, being learned using 2011 in motor electronic engineer The computer vision and pattern of meeting (Institute of Electrical and Electronics Engineers, IEEE) are distinguished Page 345-352 of knowledge journal (computer vision and pattern recognition, CVPR) is delivered " flat The method that the sliding affine splicing (Smoothly Varying Affine Stitching) of variation " paper is proposed coincides to obtain Image.
Later, in step S545, whether image is complete after judging adjustment, such as image whether there is sky after judgement adjustment White region or defect area.If image is complete after adjustment, as shown in step S550, image charts after being after storage adjustment Frame.If image is imperfect (such as there are white space or defect areas) after adjustment, as shown in step S555, then takes out and mend The adjacent another picture frame different from picture frame N of picture frame is repaid, and continues to modify based on another picture frame.
Picture frame is compensated for example, it is assumed that taking out picture frame N-1 in step S535 and being used as, then the shadow after step S545 judgement adjustment When as being imperfect, picture frame N-2 is taken out in step S555 as next compensation picture frame with shadow after further modification adjustment Picture.If image is still imperfect after the adjustment regained, picture frame N-3 is taken out as next compensation picture frame, with such It pushes away, image is complete (that is, there is no white space or defect areas) after adjustment obtained.In another embodiment In, when image is imperfect after step S545 judgement adjustment, picture frame N+2 is taken out as next compensation picture frame further to repair Image after decorations adjustment takes out picture frame N+3 and schemes as next compensation if image still deposits imperfect after the adjustment regained Frame, and step S540~S555 is repeated, and so on.In addition, can also further set a threshold value, modified Number be more than this threshold value, can not still obtain image after complete adjustment, just abandon image after the corresponding adjustment of picture frame N, and Return step S220_1 takes out next picture frame (N=N+1) Lai Zhihang adjustment programme, until all carrying out toning to all picture frames Whole program.
Beneath the case where left side is located at the white space based on picture frame N new picture frame obtained, and with " piecing together " Mode is illustrated.Firstly, judging whether picture frame N-1 is located at the leftmost side of picture frame N based on relative positional relationship.If picture frame N- 1 is located at the leftmost side of picture frame N, obtains picture frame N-1.Then, coincide new picture frame and picture frame N-1.Then, it is with the target of picture frame N Center re-fetches image after adjustment.Later, whether image is complete after judging adjustment.That is, it is judged that image whether there is after adjustment White space.If it is determined that adjustment after image be it is complete, chart frame after image storage is after adjustment.If it is determined that image is endless after adjustment It is whole, take previous picture frame forward (previous of picture frame N-1 is picture frame N-2).Also, whether judge the previous picture frame taken out For null value, that is, whether there is previous picture frame.If being null value, adjustment picture frame N is abandoned.If not null value, then to be taken out Previous picture frame come continue with adjustment after image coincide.
If picture frame N+1 is located at the leftmost side of picture frame N, picture frame N+1 is obtained.Then, coincide new picture frame and picture frame N+1.Then, Image after adjusting is re-fetched centered on the target of picture frame N.Later, whether image is complete after judging adjustment.That is, it is judged that adjustment Image whether there is white space afterwards.If it is determined that adjustment after image be it is complete, chart frame after image storage is after adjustment.If it is determined that Image is imperfect after adjustment, removes a picture frame backward (next of picture frame N+1 is picture frame N+2).Also, what judgement was taken out Whether next picture frame is null value, that is, whether there is next picture frame.If being null value, adjustment picture frame N is abandoned.If not empty Value, then with the next picture frame taken out come continue with adjust after image coincide.
And the explanation for the case where newly white space of picture frame is located at right side is identical as left side, details are not described herein, by above-mentioned Illustrate to analogize and obtain.In other embodiments, if it is determined that image is imperfect after adjustment, other closer figures of time also be can use Frame continues to coincide, and not limiting is previous picture frame or latter picture frame.
It is beneath to illustrate again for an embodiment.Fig. 6 A~Fig. 6 D is the signal according to the modification picture frame of one embodiment of the invention Figure.Fig. 6 A, which is shown, currently takes out the picture frame 610 for being intended to be handled, and Fig. 6 B show next picture frame 620 of picture frame 610, Fig. 6 C show new picture frame 630, and Fig. 6 D show image 640 after adjustment.In the present embodiment, picture frame 620 is located at picture frame 610 Left side.Referring to Fig. 6 A and Fig. 6 C, new picture frame 630 is obtained in such a way that the target T of picture frame 610 is set to center, in new picture frame 630 left side will form white space 601.Also, image 640 after being adjusted using picture frame 620 as compensation picture frame, such as Shown in Fig. 6 D.For example, can be used the mode of " cliping and pasting " or " piecing together " come image 640 after being adjusted.
Fig. 7 is the block diagram according to the film editing system of one embodiment of the invention.Please refer to Fig. 7, film editing system 700 include electronic device 710 and calculation server 720.Electronic device 710 is, for example, smart phone, tablet computer, digital phase Machine, video camera, monitor, wearable device etc..Electronic device 710 includes network chip 711, processor 712 and storage dress Set 713.Storage device 713 includes Object selection module 714.Processor 712 is coupled to network chip 711 and storage device 713, and performance objective chooses module 714 to realize corresponding function.
Calculation server 720 is, for example, the electronic device with higher operational capability such as desktop computer, laptop.Fortune Calculating server 720 includes network chip 721, processor 722 and storage device 723.Storage device 723 includes film pre-treatment Module 724, film modification module 725 and film output module 726.Processor 722 is coupled to network chip 721 and storage Device 723, and film pre-processing module 724, film modification module 725 and film output module 726 are executed to realize pair The function of answering.
Network chip 711,721 is the component with networking function, for example, wired network adapter or wireless network card.Network chip 711,721 WiFi module, WLAN (Wireless LAN, radio area network) module, 3G (third generation are provided Of wireless mobile telecommunications technology, 3rd generation mobile communication technology) module, 4G (fourth generation of wireless mobile telecommunications technology, forth generation move The dynamic communication technology) module or LTE (Long Term Evolution, Long Term Evolution) module.
Electronic device 710 links to the network chip of network with calculation server 720 by network chip 711 721 connections, to transmit data by network.Target is selected come frame by electronic device 710, and by original film and target (target signature) is sent to calculation server 720, so that calculation server 720 carries out film editor to original film, and transmits Film is to electronic device 710 after processing.In addition, in other embodiments, also data can be transmitted by USB.
In the present embodiment, processor 712,722 is similar to processor 110, storage device 713,723 and storage device 120 is similar.Also, Object selection module 714, film pre-processing module 724, film modify module 725 and film exports mould The function of block 726 is defeated with Object selection module 121, film pre-processing module 123, film modification module 125 and film respectively Module 127 is similar out, therefore associated description can refer to above description, and details are not described herein.
As long as one can be exported by mesh by the above method in conclusion user inputs film and frame selects target Mark the film in setting.Assuming that user's input is that after the processing of this case method, it is steady to edit out the person of being taken with clapping film It is fixed motionless, and the constantly mobile stabilization of environment is with clapping film.
Although the present invention is disclosed as above with embodiment, however, it is not to limit the invention, any technical field Technical staff, without departing from the spirit and scope of the present invention, when can make a little variation and retouching, therefore protection model of the invention It encloses subject to view appended claims institute defender.

Claims (19)

1. a kind of film edit methods, comprising:
An original film is received, wherein the original film includes multiple picture frames;
A target is obtained from the original film;
Multiple image feature points based on each of picture frame, obtain each of picture frame respectively with its adjacent two institute in front and back State the relative positional relationship between picture frame;
An adjustment programme is executed to those picture frames respectively, wherein the adjustment programme includes the following steps:
Picture frame N is taken out from those picture frames;
Recognize the target in picture frame N;
Picture frame N is adjusted to obtain a new picture frame, so that the target is located at the center of the new picture frame;And
The new picture frame is modified based on the relative positional relationship and then the frame that charts after obtaining one;And
Film after a processing is exported based on multiple rear drawing frames.
2. film edit methods as described in claim 1, wherein picture frame N is adjusted to obtain the new picture frame, so that the target Include: positioned at the step of center of the new picture frame
A size according to picture frame N establishes a blank picture frame;And
By being the center of the blank picture frame by the goal-setting, and picture frame N is coincided to the blank picture frame and obtains this newly Picture frame.
3. film edit methods as claimed in claim 2, wherein modified based on the relative positional relationship the new picture frame into And before the step of obtaining the rear drawing frame, further includes:
Judge whether the new picture frame is complete, in the new incomplete situation of picture frame, this is modified based on the relative positional relationship New picture frame obtains the rear drawing frame in turn;And
In the new complete situation of picture frame, not modifying the new picture frame and storing the new picture frame is the rear drawing frame.
4. film edit methods as claimed in claim 3, wherein in the new incomplete situation of picture frame, further includes:
The position that a white space is detected where in the new picture frame;
A work close to the position is taken out from the corresponding picture frame N-1 of picture frame N and picture frame N+1 based on the relative positional relationship For a compensation picture frame;And
Based on the new picture frame and the compensation picture frame, image after an adjustment is obtained.
5. film edit methods as claimed in claim 4, wherein being somebody's turn to do where detecting the white space in the new picture frame After the step of position, further includes:
The position based on white space place in the new picture frame sets a reference side;
Judge whether picture frame N-1 and picture frame N+1 are located at the reference side of picture frame N;
If determining that picture frame N-1 or picture frame N+1 are located at the reference side of picture frame N, certainly should based on the relative positional relationship One that the close reference side is taken out in picture frame N-1 and picture frame N+1 is used as the compensation picture frame;And
If determining that picture frame N-1 and picture frame N+1 neither in the reference side of picture frame N, then abandon adjusting picture frame N.
6. film edit methods as claimed in claim 4, wherein obtaining the adjustment based on the new picture frame and the compensation picture frame The step of image includes: afterwards
According to those image feature points, the size of the white space and the position, cut and the blank area from the compensation picture frame The block is conformed to the white space of the new picture frame by the corresponding block in domain.
7. film edit methods as claimed in claim 4, wherein obtaining the adjustment based on the new picture frame and the compensation picture frame The step of image includes: afterwards
Based on those image feature points, which is coincided on the compensation picture frame and obtains one and coincides image;And
Block identical with the size of the new picture frame in the image that coincides is cut centered on the target.
8. film edit methods as claimed in claim 4, wherein obtaining the adjustment based on the new picture frame and the compensation picture frame Afterwards after the step of image, further includes:
Whether image is complete after judging the adjustment;
If image is complete after the adjustment, then image is the rear drawing frame after storing the adjustment;And
If image is imperfect after the adjustment, then another picture frame different from picture frame N adjacent with the compensation picture frame is taken out, and Continue to modify based on another picture frame.
9. film edit methods as described in claim 1, wherein those image feature points based on each of picture frame, obtain Each of picture frame respectively the relative positional relationship between its adjacent two described picture frame in front and back the step of include:
Based on those image feature points in those image feature points and picture frame N+1 in picture frame N, judge that picture frame N+1 is opposite In one first relative position of picture frame N;
Based on those image feature points in those image feature points and picture frame N-1 in picture frame N, judge that picture frame N-1 is opposite In one second relative position of picture frame N;And
First relative position and second relative position are stored as the relative positional relationship of picture frame N.
10. a kind of film editing device, comprising:
One storage device stores multiple modules, including the modification of an Object selection module, a film pre-processing module, a film Module and a film output module;And
One processor is coupled to the storage device, executes those modules to edit to an original film, the original film packet Multiple picture frames are included,
Wherein, which receives the original film, and obtains a target from the original film;
Multiple image feature points of the film pre-processing module based on each of picture frame, obtain each of picture frame respectively with its Relative positional relationship between two adjacent picture frames of front and back;
The film modifies module and executes an adjustment programme to those picture frames respectively, which includes: to take from those picture frames Picture frame N out;Recognize the target in picture frame N;Picture frame N is adjusted to obtain a new picture frame, so that the target is located at the new figure The center of frame;And the new picture frame is modified based on the relative positional relationship and then the frame that charts after obtaining one;
The film output module exports film after a processing based on multiple rear drawing frames.
11. film editing device as claimed in claim 10, wherein film modification module is built according to a size of picture frame N A vertical blank picture frame;And by by the goal-setting being the center of the blank picture frame, and picture frame N is coincided to the blank sheet Frame and obtain the new picture frame.
12. film editing device as claimed in claim 11, wherein film modification module judges whether the new picture frame is complete, In the new incomplete situation of picture frame, the new picture frame is modified based on the relative positional relationship and then obtains the rear drawing frame; In the new complete situation of picture frame, not modifying the new picture frame and storing the new picture frame is the rear drawing frame.
13. film editing device as claimed in claim 12, wherein it is new in this to detect a white space for film modification module A position where in picture frame;It is taken out and is leaned on from picture frame N corresponding picture frame N-1 and picture frame N+1 based on the relative positional relationship One of the nearly position is as a compensation picture frame;And based on the new picture frame and the compensation picture frame, obtain image after an adjustment.
14. film editing device as claimed in claim 13, wherein it is new in this to be based on the white space for film modification module The position where in picture frame sets a reference side;Judge whether picture frame N-1 and picture frame N+1 are located at the ginseng of picture frame N Examine side;If determining that picture frame N-1 or picture frame N+1 are located at the reference side of picture frame N, certainly should based on the relative positional relationship One that the close reference side is taken out in picture frame N-1 and picture frame N+1 is used as the compensation picture frame;If determine picture frame N-1 with Picture frame N+1 then abandons adjusting picture frame N neither in the reference side of picture frame N.
15. film editing device as claimed in claim 13, wherein film modification module according to those image feature points, should The size of white space and the position cut block corresponding with the white space from the compensation picture frame, which are bonded To the white space of the new picture frame.
It, will 16. film editing device as claimed in claim 13, wherein film modification module is based on those image feature points The new picture frame coincides on the compensation picture frame and obtains one and coincide image;And it is cut centered on the target in the image that coincides Block identical with the size of the new picture frame.
17. film editing device as claimed in claim 10, wherein film modification module judges after the adjustment whether is image Completely;If image is complete after the adjustment, then image is the rear drawing frame after storing the adjustment;If image is endless after the adjustment It is whole, then another picture frame different from picture frame N adjacent with the compensation picture frame is taken out, and continue based on another picture frame Modification.
18. film editing device as claimed in claim 10, wherein the film pre-processing module is based on those in picture frame N Those image feature points in image feature point and picture frame N+1 judge picture frame N+1 relative to the one first opposite of picture frame N Position;Based on those image feature points in those image feature points and picture frame N-1 in picture frame N, picture frame N-1 phase is judged For one second relative position of picture frame N;And first relative position and second relative position are stored as picture frame N's The relative positional relationship.
19. a kind of film editing system, comprising:
One calculation server;And
One electronic device, comprising:
One first network chip;
One first storage device stores an Object selection module;And
One first processor is coupled to first storage device, executes the Object selection module, is connect by the Object selection module Packet receiving includes an original film of multiple picture frames, and obtains a target from original film, transmits the original by the first network chip Beginning film and the target are to the calculation server, so that the calculation server carries out a film editor to the original film, and Film is to the electronic device after one processing of transmission;
Wherein, the calculation server, comprising:
One second network chip receives the original film and the target from the electronic device by a network;
One second storage device stores multiple modules, including a film pre-processing module, a film modification module and a shadow Piece output module;And
One second processor is coupled to the storage device, drives those modules to carry out film editor,
Wherein, multiple image feature points of the film pre-processing module based on each of picture frame obtain each of picture frame point Relative positional relationship not between two picture frames adjacent with its front and back;
The film modifies module and executes an adjustment programme to those picture frames respectively, which includes: to take from those picture frames Picture frame N out;Recognize the target in picture frame N;Picture frame N is adjusted to obtain a new picture frame, so that the target is located at the new figure The center of frame;And the new picture frame is modified based on the relative positional relationship and then the frame that charts after obtaining one;
The film output module exports film after the processing based on multiple rear drawing frames.
CN201710874533.6A 2017-09-13 2017-09-25 Video editing method, device and system Active CN109495778B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
TW106131373A TWI622018B (en) 2017-09-13 2017-09-13 Method, device and system for editing video
TW106131373 2017-09-13

Publications (2)

Publication Number Publication Date
CN109495778A true CN109495778A (en) 2019-03-19
CN109495778B CN109495778B (en) 2021-09-28

Family

ID=62640026

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710874533.6A Active CN109495778B (en) 2017-09-13 2017-09-25 Video editing method, device and system

Country Status (3)

Country Link
US (1) US10360943B2 (en)
CN (1) CN109495778B (en)
TW (1) TWI622018B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2019105933A (en) * 2017-12-11 2019-06-27 キヤノン株式会社 Image processing apparatus, method of controlling image processing apparatus, and program
CN112449229B (en) * 2019-09-04 2022-01-28 海信视像科技股份有限公司 Sound and picture synchronous processing method and display equipment
CN114245169B (en) * 2021-12-21 2024-01-23 北京达佳互联信息技术有限公司 Video processing method, apparatus, electronic device, medium, and program product

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1766929A (en) * 2004-10-29 2006-05-03 中国科学院计算技术研究所 A kind of motion object motion reconstructing method based on three-dimensional data base
CN101340513A (en) * 2007-07-02 2009-01-07 华硕电脑股份有限公司 Image processing method and electronic device applying same
US20100225773A1 (en) * 2009-03-09 2010-09-09 Apple Inc. Systems and methods for centering a photograph without viewing a preview of the photograph
CN104252705A (en) * 2014-09-30 2014-12-31 中安消技术有限公司 Method and device for splicing images
CN104301630A (en) * 2014-09-10 2015-01-21 天津航天中为数据系统科技有限公司 A video image splicing method and device
CN105979383A (en) * 2016-06-03 2016-09-28 北京小米移动软件有限公司 Image acquisition method and device

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5973732A (en) * 1997-02-19 1999-10-26 Guthrie; Thomas C. Object tracking system for monitoring a controlled space
EP2100453B1 (en) * 2006-12-18 2013-07-03 FUJIFILM Corporation Monitoring system, monitoring method and program
US9019381B2 (en) * 2008-05-09 2015-04-28 Intuvision Inc. Video tracking systems and methods employing cognitive vision
TWI384408B (en) * 2009-04-30 2013-02-01 Ind Tech Res Inst Method and system for identifying image and outputting identification result
TWI381321B (en) * 2009-04-30 2013-01-01 Ind Tech Res Inst Method for image recombination of multiple images and identifying image and system for identifying image and outputting identification result
US9740895B1 (en) * 2014-05-30 2017-08-22 Google Inc. Method and system for identifying and tracking tagged, physical objects
US20170244959A1 (en) * 2016-02-19 2017-08-24 Adobe Systems Incorporated Selecting a View of a Multi-View Video

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1766929A (en) * 2004-10-29 2006-05-03 中国科学院计算技术研究所 A kind of motion object motion reconstructing method based on three-dimensional data base
CN101340513A (en) * 2007-07-02 2009-01-07 华硕电脑股份有限公司 Image processing method and electronic device applying same
US20100225773A1 (en) * 2009-03-09 2010-09-09 Apple Inc. Systems and methods for centering a photograph without viewing a preview of the photograph
CN104301630A (en) * 2014-09-10 2015-01-21 天津航天中为数据系统科技有限公司 A video image splicing method and device
CN104252705A (en) * 2014-09-30 2014-12-31 中安消技术有限公司 Method and device for splicing images
CN105979383A (en) * 2016-06-03 2016-09-28 北京小米移动软件有限公司 Image acquisition method and device

Also Published As

Publication number Publication date
CN109495778B (en) 2021-09-28
TW201915939A (en) 2019-04-16
TWI622018B (en) 2018-04-21
US20190080718A1 (en) 2019-03-14
US10360943B2 (en) 2019-07-23

Similar Documents

Publication Publication Date Title
US11538232B2 (en) Tracker assisted image capture
EP3683572B1 (en) Battery detection method and device
US8363933B2 (en) Image identification method and imaging apparatus
CN107886484B (en) Beautifying method, beautifying device, computer-readable storage medium and electronic equipment
CN106101561B (en) Camera focusing detection method and device
CN106899781B (en) Image processing method and electronic equipment
US11343591B2 (en) Method and system of presenting moving images or videos corresponding to still images
US9992408B2 (en) Photographing processing method, device and computer storage medium
WO2014116485A1 (en) Systems and methods for tracking and detecting a target object
EP2709062A2 (en) Image processing device, image processing method, and computer readable medium
CN114302226B (en) Intelligent cutting method for video picture
CN101299239B (en) Method and device for acquiring character area image and character recognition system
CN109495778A (en) film editing method, device and system
CN114022823A (en) Shielding-driven pedestrian re-identification method and system and storable medium
CN109151325A (en) A kind of processing method and processing unit synthesizing smiling face
CN110266955A (en) Image processing method, device, electronic device, and storage medium
CN110213486A (en) Image capturing method, terminal and computer readable storage medium
CN111476056B (en) Target object identification method, device, terminal equipment and computer storage medium
CN103929585B (en) A kind of control method of polaroid, electronic equipment and system
CN113591865A (en) Loop detection method and device and electronic equipment
CN113132638B (en) Video processing method, video processing system, mobile terminal and readable storage medium
WO2023193648A1 (en) Image processing method and apparatus, electronic device, and storage medium
CN114972801B (en) Video generation method, device, equipment and readable storage medium
CN117437122B (en) Method and system for splicing panoramic images of container
US9135687B2 (en) Threshold setting apparatus, threshold setting method and recording medium in which program for threshold setting method is stored

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant